The Architecture of Agency Volume 7 Words That Lost Their Teeth

Words That Lost Their Teeth

Semantic hygiene for political language

This chapter is a review — it is readable but still changing.

Consider what the word “socialist” is asked to cover. It names the regime that built Auschwitz. It names the regime that built the Gulag. And it names the regime that built the Danish healthcare system. One word, stretched across an ethno-nationalist death cult, a class-revolutionary police state, and a prosperous parliamentary democracy where the trains run on time and nobody disappears in the night. A word that covers all three of those things conveys almost nothing. Whoever uses it unqualified is not making a claim; he is making a noise.

“Fascist” has likewise eroded from ultranationalism, corporatist statism, militarism, and violent authoritarian control into “people I dislike on the right.” The same wear has claimed “communist,” “liberal,” “neoliberal,” and “far right.” Instruments became gestures.

This is more than pedantry. Political vocabulary is an instrument panel; gauges that always read the same leave judgment blind. A society unable to distinguish a Nazi from a social democrat in language will condemn the harmless and excuse the monstrous.

The remedy is not new words. It is taxonomy plus discipline: split the inflated terms into the distinct things they name, then hold the line on usage. This chapter does that for the two worst offenders.

Three Socialisms

The word socialism has historically named three systems with different ideals, methods, and outcomes.

National Socialism is ethno-nationalist authoritarianism. Nazi Germany appropriated worker and anti-capitalist language while subordinating freedom and economic rights to the nation-state and racial identity. Its economics were corporatist: private ownership on paper, state direction in fact. Oppression, violence, and genocide define its legacy.

International Socialism — Marxist-Leninist communism — championed global working-class solidarity and abolition of private ownership. The Soviet Union and Maoist China centralized political and economic command, crushing individual rights in the name of class struggle. Its record is broadly discredited, though it retains Western admirers who never lived under it.

Liberal Socialism — social democracy or welfare capitalism — keeps democratic government, individual rights, private ownership, and markets while adding regulation and social provision. Nordic countries, Canada, and much of Western Europe have made this model stable and durably popular.

Set the three side by side and the differences stop being subtle:

Aspect National Socialism International Socialism Liberal Socialism
Central goal Ethnic-nationalist supremacy International proletarian solidarity Democratic individual and collective flourishing
Economic control Corporatist authoritarianism Centrally planned economy Regulated market economy
Individual freedom Severely restricted Severely restricted historically Protected and prioritized
Democracy Explicitly anti-democratic Explicitly anti-democratic historically Explicitly democratic

The Axis That Matters

On the conventional axis, National and International Socialism are mortal opposites — far right and far left, enemies from Weimar’s streets through the Eastern Front. Yet on every row measuring treatment of people, they converge: authoritarian government, suppressed rights, and state control. The rivalry was real; the resemblance is realer. Two prison camps do not become opposites because their flags differ.

Liberal Socialism sits nearer International Socialism in its language of solidarity and redistribution, yet is the structural outlier: rights protected, democracy explicit, markets retained.

The structural axis — how power is held and constrained — matters more than what power is nominally for. The first two systems name different beneficiaries, race and class, but share a total state, blocked exit, and punished dissent. The third shares some distributive goals while rejecting the machinery. Content divides the first two; structure unites them and separates both from Liberal Socialism.

This is one of two complementary lenses this book applies to socialism: this chapter’s taxonomy sorts regime types by structure, while the dose taxonomy of Against Socialism — dose, domain, monopoly, exit, feedback — measures how much of the socialist mechanism a system administers and how it is constrained; the two lenses answer different questions and neither substitutes for the other. And it is the same insight that the political triangle in Why Liberalism Loses encodes geometrically: the interesting divide in politics is not left versus right but the axis running orthogonal to it.

Why Precision Matters

Everybody knows Sweden is not the Soviet Union, but arguments run on words. A term binding three referents lets a conclusion proved about one travel silently to the others.

Call a national healthcare proposal “socialism” and you have implied — without ever having to argue it — that it sits on a continuum ending in gulags. Call it that in front of an audience that associates the word with Denmark and you have implied it is obviously benign. Both moves are fraudulent, and both are ubiquitous. The defender of welfare capitalism gets saddled with Stalin’s body count; the apologist for central planning hides behind Norwegian pension funds. Each equivocation is only possible because the word refuses to distinguish.

The confusion compounds when inflated words collide: “The Nazis were socialists, so the left are the real fascists”; “Fascism is capitalism in decay, so the right are the real fascists.” Terms baggy enough to hold any opponent prevent convergence because the parties no longer discuss the same object. Distinguishing the three socialisms is a precondition for arguing about them.

Doyle’s Razor

“Fascist” is the most inflated label. When everyone is a fascist, the word means nothing and genuine fascists hide among the falsely accused.

The writer Andrew Doyle has proposed a simple intervention, which I will call Doyle’s razor:

If you condone the murder of your political opponents, you’re a fascist.

This is not exhaustive — historical fascism includes more than political murder — but it is a sharp sufficient condition, cheap to check and indifferent to faction.

The test earns its keep whenever political violence occurs. After a political assassination in a polarized society, the pattern is depressingly reliable: the great majority across the spectrum condemn the killing, while a fringe openly celebrates it — and screenshots of the celebrants circulate until they create the impression that a large fraction of the other side was cheering. That impression is exaggeration, and the exaggeration is itself a form of the inflation this chapter is against. What exists is a noisy fringe: individuals who crossed the line into condoning political murder.

Doyle’s razor preserves both judgments. The celebratory fringe are not their movement’s mainstream, but cheering political bloodshed crosses the line. Fascism is not a neighbor who votes wrong or holds an unpopular opinion; here it is the legitimization of political murder, whatever banner covers it.

Notice that the discourse around fascism suffers from two opposite distortions, and the razor slices through both:

  1. Overuse. Calling every ordinary conservative — or every ordinary progressive — a fascist drains the term of meaning and slanders the innocent.
  2. Under-focus. Ignoring the fringe that condones political violence leaves genuine fascism unnamed and uncontested.

The errors reinforce each other. Overuse turns accusation into noise and gives the guilty cover; under-focus leaves them unnamed. Moral condemnation is currency: debase it and it buys nothing. Doyle’s test preserves weight through precision.

I should be clear about what this chapter is and is not claiming. The inflation of political labels is not always innocent drift; often it is deliberate — a weapon in memetic conflict, wielded to destroy reputations and forbid positions rather than describe them. The mechanics of that weaponization — how cancel culture deploys inflated labels as social artillery, and how algorithmic media corrupt language itself — belong to the next volume’s treatment of narrative warfare. This chapter’s business is the definitional hygiene that makes the weapon detectable: you cannot recognize a debased currency until you know what the coin was originally worth.

A LARPitarian Interlude

Semantic hygiene has a temperamental cousin: hold political identity lightly. Erving Goffman described social life as contextual performance, and political flags, team language, and ritual denunciations are often costume. Call the honest version LARPitarianism: everybody should be free to LARP; nobody should be forced to LARP.

The first clause protects the right to adopt and discard roles, including ridiculous ones. The second forbids compulsory identities, rituals, and casting strangers as villains. Label inflation is forced LARP: a casting director nobody hired conscripts the public as extras.

Retiring the Compressed Axes

“Socialist,” “fascist,” and left–right all compress high-dimensional political arrangements into an epithet or line. The table shows the loss: structural twins land at opposite ends and structural opposites side by side. Equivocations and label warfare exploit that loss.

Taxonomy and discipline are the repair this chapter can offer: split the words, publish the table, apply the razor. But the honest endpoint of the argument is that the compressed axes should not be repaired so much as retired. What politics actually needs is more dimensions, not better epithets — a way of locating political arrangements in the full space of their structural properties: how power is held, whether exit exists, what feedback constrains the rulers, how much of life the system claims. In that space, “how far left?” dissolves as a question, and the question that replaces it — is this arrangement inside or outside the region where agency remains viable? — turns out to be answerable. Building that space is the work of The Admissible Region. This chapter has only sharpened the old tools; that one replaces them.